Posted on 07/21/2009 5:31:52 PM PDT by SeekAndFind
One of the challenges of working in the IT field is staying on top of emerging technologies - while letting go of those that are becoming obsolete. This Global Knowledge article lists 10 areas that are fading into obscurity.
There are some things in life, like good manners, that never go out of style. And there are other things, like clothing styles, that fall in and out of fashion. But when an IT skill falls out of favor, it rarely ever comes back. Heres our list of 10 dying IT skills. If any of these skills is your main expertise, perhaps its time to think about updating your skill set.
1: Asynchronous Transfer Mode
ATM was popular in the late 90s, particularly among carriers, as the answer to overworked frame relay for wide-area networking. It was considered more scalable than frame relay and offered inherent QoS support. It was also marketed as a LAN platform, but that was its weakness. According to Wikipedia, ATM failed to gain wide acceptance in the LAN where IP makes more sense for unifying voice and data on the network. Wikipedia notes that ATM will continue to be deployed by carriers that have committed to existing ATM deployments, but the technology is increasingly challenged by speed and traffic shaping requirements of converged voice and data networks. A growing number of carriers are now using Multi-Protocol Label Switching (MPLS), which integrates the label-switching capabilities of ATM with the packet orientation of IP. IT skills researcher Foote Partners listed ATM in its IT Skills and Certification Pay Index as a non-certified IT skill that has decreased in value in the last six month of 2008.
2: Novell NetWare
Novells network operating system was the de facto standard for LANs in the 1990s, running on more than 70% of enterprise networks. But Novell failed to compete with the marketing might of Microsoft. Novell tried to put up a good fight by acquiring WordPerfect to compete with Windows Office, but that move failed to ignite the market, and Novell eventually sold WordPerfect to Corel in 1996. Novell certifications, such as Certified Novell Engineer, Master Certified Novell Engineer, Certified Novell Certified Directory Engineer, and Novell Administrator, were once hot in the industry. But now, they are featured in Foote Partners list of skills that decreased in value in 2008. Hiring managers want Windows Server and Linux skills instead.
3: Visual J++
Skills pay for Microsofts version of Java declined 37.5% last year, according to the Foote Partners study. The life of J++, which is available with Microsoft Visual Studio 6.0, was not a smooth one. Although Sun Microsystems licensed Java to Microsoft to develop J++, Microsoft failed to implement some features of the official Java standard while implementing other extensions of its own. Sun sued Microsoft for licensing violations in a legal wrangle that lasted three years. Microsoft eventually replaced J++ with Microsoft .NET.
4: Wireless Application Protocol
Yes, people were able to browse the Internet in the late 90s before Apples iPhone. Web site operators would rewrite their content to the WAPs Wireless Markup Language, enabling users to access Web services such as email, stock results and news headlines using their cell phones and PDAs. WAP was not well received at the beginning because WAP sites were slow and lacked the richness of the Web. WAP has also seen different levels of uptake worldwide because of the different wireless regulations and standards around the world. WAP has since evolved and is a feature of Multimedia Messaging Service, but there is now a new generation of competing mobile Web browsers, including Opera Mobile and the iPhones Safari browser.
5: ColdFusion
ColdFusion users rave that this Web programming language is easy to use and quick to jump into, but as many other independent software tools have experienced, its hard to compete with products backed by expensive marketing campaigns from Microsoft and others. The language was originally released in 1995 by Allaire, which was acquired by Macromedia (which itself was purchased by Adobe). Today, it is superseded by Microsoft .NET, Java, PHP, and the language of the moment: open source Ruby on Rails. A quick search of the Indeed.com job aggregator site returned 11,045 jobs seeking PHP skills, compared to 2,027 CF jobs. Even Ruby on Rails, which is a much newer technology - and which received a major boost when Apple packaged it with OS X v10.5 in 2007 returned 1,550 jobs openings on Indeed.com.
6: RAD/extreme programming
Back in the late 90s and early 2000s, the rapid application development and extreme programming development philosophies resulted in quicker and more flexible programming that embraced the ever-changing needs of customers during the development process. In XP, developers adapted to changing requirements at any point during the project life rather than attempting to define all requirements at the beginning. In RAD, developers embraced interactive use of structured techniques and prototyping to define users requirements. The result was accelerated software development. Although the skills were consistently the highest paying in Foote Partners survey since 1999, they began to lose ground in 2003 due to the proliferation of offshore outsourcing of applications development.
7: Siebel
Siebel is one skill that makes a recurring appearance in the Foote Partners list of skills that have lost their luster. Siebel was synonymous with customer relationship management in the late 90s and early 2000s, and the company dominated the market with a 45% share in 2002. Founded by Thomas Siebel, a former Oracle executive with no love lost for his past employer, Siebel competed aggressively with Oracle until 2006 when it was ultimately acquired by the database giant. Siebels complex and expensive CRM software required experts to install and manage. That model lost out to the new breed of software-as-a-service (SaaS) packages from companies such as Salesforce.com, which deliver comparable software over the Web. According to the ITJobsWatch.com, Siebel experts command an average salary of GBP52,684 ($78,564), but thats a slide from GBP55,122 a year ago. Siebel is ranked 319 in the job research sites list of jobs in demand, compared to 310 in 2008.
8: SNA
The introduction of IP and other Internet networking technologies into enterprises in the 1990s signaled the demise of IBMs proprietary Systems Network Architecture. According to Wikipedia, the protocol is still used extensively in banks and other financial transaction networks and so SNA skills continue to appear in job ads. But permanent positions seeking SNA skills are few and far between. ITJobsWatch.com noted that there were three opening for permanent jobs between February and April, compared to 43 during the same period last year. Meanwhile, companies such as HP offer consultants with experience in SNA and other legacy skills, such as OpenVMS and Tru64 UNIX for short-term assignments.
9: HTML
Were not suggesting the Internet is dead, but with the proliferation of easy-to-use WYSIWYG HTML editors enabling non-techies to set up blogs and Web pages, Web site development is no longer a black art. Sure, theres still a need for professional Web developers, but a good grasp of HTML isnt the only skill required of a Web developer. Professional developers often have expertise in Java, AJAX, C++, and .NET, among other programming languages. HTML as a skill lost more than 40% of its value between 2001 and 2003, according to Foote Partners.
10: COBOL
Is it dead or alive? This 40-year-old programming language often appears in lists of dying IT skills. But it also appears in as many articles about organizations with legacy applications written in COBOL that are having a hard time finding workers with COBOL skills. IBM cites statistics that 70% of the worlds business data is still being processed by COBOL applications. But how many of these applications will remain in COBOL for the long term? Even IBM is pushing its customers to build bridges and use service-oriented architecture to transform legacy applications and make them part of a fast and flexible IT architecture. About the author
--------------------------------------------------------------------------------------------------------------
Linda Leung is a senior IT journalist with 20 years experience editing and writing news and features for online and print. She has extensive experience creating and launching news Web sites, including most recently, independent communities for customers of Cisco Systems and Microsoft.
With all the COBOL love flowing here, let’s take a second to remember that great comptuer pioneer, Admiral Grace Hopper, the mother of COBOL (and finder of the first literal computer bug and the one who first implemented language standards). It may look antiquated now, but it was revolutionary in the 60s.
COBOL if it still works, why fix it.
What do you guys think about Visual Basic for Applications (VBA)?
Very true. I work for a small section of the Dept of Education that is using such an old legacy system that there is only one (very old) employee left in the whole nation who knows the programming (”Easytrieve”).
When he retires I don’t know what we’ll do. No contractor wants to get involved with such old boring stuff, and no federal employees can seem to learn to program.
We can’t get a new system because when we try to get new huge expensive contracts to do so, they somehow fall apart and fail.
I guess that also leaves out anyone who can re-punch an employee’s payroll information on a card when the damn reader eats it? (sarc. extreme)
I had a buddy who received a panic call about some machine that he had done that had some bug that couldn't be described very well but which it was imperative had to be fixed immediately. So he gets on a plane, goes to the plant and checks in. He's sent out to the machine to check it out. Well, he operates it a few cycles and can't seem to find anything out of order. Tells the powers that be, who say they'll send out the operator to demo the problem. So the guy comes out and he presses several buttons on the control panel simultaneously, while reaching way out with his foot to trip some sensor in the machine, causing the machine to malfunction in some way. That was the "bug".
I knew another guy who was on site commissioning some equipment which was functional enough to use and needed to be. So during the day, he would do software things he could do online (the platform allows online edits without a restart), then after hours do more invasive things he couldn't do while running. So he comes in the next morning and is told the machine won't start. Goes over to the machine, presses "Start" on the touchscreen, and wham-o, magick-o it starts. Turns out the operator can't read (!) and has learned to operate the touchscreen by remembering button positions. The guy had moved the location of the start button the previous night to make room on the screen for some added features.
So anytime you think you've constructed a foolproof system, the universe simply starts supplying a better grade of fools.
Old school data processing.....those were the days!....Mag tapes dumping in the vacume column...cards getting mangled in the reader. That damn ball and string on the master console breaking during an IPL....Many bars were filled with D.P. people after hours.....
So long as the COBOL systems function, can be updated at moderate/reasonable costs, and developers/maintainers can be found, COBOL will continue to thrive. There’s just too much vested in those systems to risk making a multi-billion-dollar transition to a new system which does exactly the same thing.
One fact of programming: legacy systems embody a great deal of wisdom which is not otherwise documented or known. Wholesale replacement of a system is a tremendous risk precisely because those creating the new system don’t know what vital yet obscure processes are performed by the old. Such replacement rarely happens unless there is an imperative to do so, such as outright parts obsolescence or skyrocketing cost of maintenance. In the case of big-iron legacy systems, it’s cheaper to teach someone an old language.
Kinda like Latin: allegedly dead, yet there is still great value in learning and using it in narrow yet big-budget “systems”.
Gracie Hopper! (Doffs hat, hand over heart). Ur-Geek before being a geek was cool and even before being a geek wasn’t cool.
The latest incarnation is Object-Oriented Cobol for the Java Virtual Machine. Much as we want it to be, and as terrifying as that concept is, COBOL isn't dead yet.
If it finally dies, methinks it will be because no self-respecting programmer will learn it.
Live, yes. Thrive, no.
The world is OOP now for lots of good reasons.
Maybe, but not dead. Seen just a few days ago:
Recent standout demo 'Pimp My Spectrum' a collection of complex effects and in-jokes wrapped in comforting Spectrum colour-clash essentially involved creator Ate Bit redesigning the 8-bit computer on the fly. "Technically, it was fairly straightforward. The hard bit was coming up with the concept, story-boarding it, getting the art and music assets created and then putting it all together before the deadline," says coder Paul Grenfel. "I mean, it's got a simple Z80 emulator in there and a load of Z80 code to run the demo, as well as a software rasteriser for the 3D and an AY emulator, but none of those were that tricky to write." The demo, technically a 64kB PC intro, sticks to its Z80 principles but shows off effects that wouldn't have been considered possible on the venerable processor in its heydey.OK, so it's really obscure - but it ain't dead.
(BTW: I wonder what the going price is on my old and nearly perfect copy of "Captain Zilog" comic book is...)
I worked on those. My recollection is that, while the decimal and floating point instruction sets were extra-cost options, all S/360s had index registers. However, not all S/360 instructions could use them. For instance, the Move Immediate instruction could not use an index register because the right half of the immediate byte took up the bit field that would have specified which index register to use.
Also, the assembler had a glitch with regard to index registers. If you were not using an index register, and you forgot to put a comma before your intended base register, the assembler would output an instruction that had an index register but no base register. Your program would still work, but would be slower because the CPU would waste an extra cycle adding in zero. Someone I knew wrote a utility program in assembly language to go through old source code and fix instances where coders had made that mistake.
On one hand you have users who can crash a system.
You haven’t been doing your job if they can do that in normal use. But even if they’re being stupid you should not let them crash it. The monkey test on the keyboard (or buttons) is a popular QA test. The meaner QA people will literally unplug your database or application server in the middle of something. But the meaner QA means less chance of a user being able to screw it up. You do have to protect them from themselves to some extent.
On the start button, good UI principles can erase a lot of user complaints, but at some point you just can’t overcome a user’s stupidity. I’ve seen someone printing over and over and over and complaining the printer’s not working. He had the Microsoft XPS writer set as his default printer. You can’t really do anything about that but try to educate them.
Except on a Mac you see a picture of the printer you’re printing to, so he may have figured out a similiar situation on a Mac by himself.
I always had the attitude that QA is helping me make my program better. In the end I get the credit for writing the program anyway, and QA just helps make me look good. Nobody ever thanks QA.
And COND codes.
-PJ
I empathize. C was developed to make it easier to develop the UNIX operating system. Then it started to get used to develop everything, and it wasn't pretty.
What I always hate is after ship when the customers run into something then everybody starts asking how QA missed that, without bothering to notice that QA was writing 5 bugs per person per day for the whole project. How’d we miss it, we were busy with other bugs.
Oh well, that’s life. It’s nice when the bugs out in the field are minor.
I would say "Monkey hit the button," but apparently your humans can't even find the button. They'll always build a better idiot.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.